© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 22-TRANSCODER-HP
    7. 110936
    Prev
    Next
    INDEX
    Explanations

    numbers and percentages

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_22.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.22.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
    çķħ
    -0.28
    erable
    -0.27
    æŃ
    -0.26
    ãĥıãĥ³
    -0.26
    è·Łè¸ª
    -0.25
    gable
    -0.25
    ĥģ
    -0.25
    rowse
    -0.25
    åĽ´
    -0.24
     hunters
    -0.24
    POSITIVE LOGITS
     posing
    0.29
    å¹´ä¹ĭä¹ħ
    0.26
     nad
    0.26
     Known
    0.26
    Known
    0.26
     poses
    0.26
     known
    0.26
     Trá»įng
    0.25
    楫
    0.25
     pose
    0.24
    Activations Density 0.024%

    No Known Activations